paper-with-me

홈 › Papers

Automatic Classification of Magnetic Chirality of Solar Filaments from H-Alpha Observations

2025-09-21 · Alexis Chalmers, Azim Ahmadzadeh arxiv

In this study, we classify the magnetic chirality of solar filaments from H-Alpha observations using state-of-the-art image classification models. We establish the first reproducible baseline for solar filament chirality classification on the MAGFiLO dataset. The MAGFiLO dataset contains over 10,000 manually-annotated filaments from GONG H-Alpha observations, making it the largest dataset for filament detection and classification to date. Prior studies relied on much smaller datasets, which limited their generalizability and comparability. We fine-tuned several pre-trained, image classification architectures, including ResNet, WideResNet, ResNeXt, and ConvNeXt, and also applied data augmentation and per-class loss weights to optimize the models. Our best model, ConvNeXtBase, achieves a per-class accuracy of 0.69 for left chirality filaments and $0.73$ for right chirality filaments.

📄 PDF Abstract BibTeX arXiv:2509.18214

Code (0)

등록된 구현이 없습니다.

Tasks

Image ClassificationData Augmentation

Similar Papers 제목 키워드 기반

Machine learning for reconstruction of polarity inversion lines from solar filaments

2024-05-10 · V. Kisielius, E. Illarionov

Solar filaments are well-known tracers of polarity inversion lines that separate two opposite magnetic polarities on the solar photosphere. Because observations of filaments began long before the systematic observations …

Super-Resolution

Chirality-Induced Magnetization of Magnetite by an RNA Precursor

2023-04-13 · S. Furkan Ozturk, Deb Kumar Bhowmick, Yael Kapon, Yutao Sang 외

Life is homochiral and homochirality is a fundamental feature of living systems on Earth. While the exact mechanism that led to homochirality is still not fully understood, any realistic scenario on the origins of life n…

Variational autoencoders understand knot topology

2025-04-05 · Anna Braghetto, Sumanta Kundu, Marco Baiesi, Enzo Orlandini

Supervised machine learning (ML) methods are emerging as valid alternatives to standard mathematical methods for identifying knots in long, collapsed polymers. Here, we introduce a hybrid supervised/unsupervised ML appro…

Decodervalid

Solar Filaments Detection using Active Contours Without Edges

2024-12-30 · Sanmoy Bandyopadhyay, Vaibhav Pant

In this article, an active contours without edges (ACWE)-based algorithm has been proposed for the detection of solar filaments in H-alpha full-disk solar images. The overall algorithm consists of three main steps of ima…

Image SegmentationObjectobject-detectionObject Detection+1

EdgeAttNet: Towards Barb-Aware Filament Segmentation

2025-09-03 · Victor Solomon, Piet Martens, Jingyu Liu, Rafal Angryk arxiv

Accurate segmentation of solar filaments in H-alpha observations is critical for determining filament chirality, a key factor in the behavior of Coronal Mass Ejections (CMEs). However, existing methods often fail to capt…